In recent years the increasing prevalence of applications for mobile devices has led to greater amounts, and more varied amounts, of data traffic communicated through networks and a greater load on the platforms that support these applications. Further, different applications may have different usage profiles, with certain applications having large surges of usages during particular time periods but relatively little usage at other times. Other applications may have a relative constant usage.
Platforms that support mobile applications may, however, have a finite capacity. This capacity may be exceeded, especially in times of high demand. However, it is an inefficient and perhaps costly use of resources to deploy additional platforms to support peak usage of the application if those platforms will be underutilized once the peak demand subsides.
Accordingly, it is desired to dynamically and efficiently deploy and deprecate platforms for mobile applications and dynamically route those mobile applications to the various deployed platforms to achieve a variety of desired goals.